I am doubting how to feed data (a set of urls, addresses, and prices) to my program, which will run periodically. There are X workflows and one of them will run each time depending on which URL it is given. It will input Y addresses into that URL and for each address it will input Z prices.
Queues and transactions
Originally I imagined a spreadsheet with a structure such as:
URL Address Price
url_1 address_1 price_1
url_1 address_1 price_2
url_1 address_1 price_z
url_2 address_2 price_1
url_2 address_2 price_2
url_2 address_2 price_z
url_2 address_y price_1
url_2 address_y price_z
url_x address_1 price_1
url_x address_2 price_2
url_x address_y price_z
I hope that makes the structure clear. Essentially I’m doing every combination of url, address, and price possible. This could all go into an Excel sheet but it doesn’t seem very maintainable. In this scenario I understand these could be transaction items that are added to a queue periodically and consumed.
Arrays in spreadsheet
The other option is to have a sheet that just looks like
urls url_1, url_2, …, url_x
addresses address_1, address_2, …, address_y
prices price_1, price_2, …, price_z
Each array could be either a string (to be split inside the program into its array items) or each cell in that row occupying an array item. My program would then just use 3 for loops to combine them as the program executes.
The final option I’m considering is similar to the above, but the three variables are stored as assets in Orchestrator. My program would use them as mentioned in the previous option.
Which of the 3 options would be considered best practice? I’ve been reading posts like this one but I’m not able to figure out my situation. Of course, if there is a better, 4th option, I’d like to hear it.